Redundancy of Structural Systems in The Context of Structural Safety

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Available online at www.sciencedirect.

com

Procedia Engineering 14 (2011) 2172–2178

The Twelfth East Asia-Pacific Conference on Structural Engineering and Construction

Redundancy of Structural Systems in the Context of


Structural Safety
Z. X. FANG a*, H. T. FAN a
a
School of Civil Engineering, Yantai University, China

Abstract

Redundancy has long been regarded as a desirable property in ensuring the safety of structural systems.
Nevertheless, it should be remarked that there is, as yet, neither a uniform theory of structural redundancy nor any
widely agreed definitions. With the aim of incorporating redundancy into safety decision-making, the present article
concentrates on the essentials of redundancy as well as its significance within the wider context of structural safety.
First, it explains three terms associated with redundancy, including static indeterminacy, alternative load paths and,
meanwhile, clarifies their interdependence. Secondly, different levels of structural safety (i.e. reliability and
robustness) are investigated from a broader perspective, in which reliability refers to the pristine or intact state of a
structure, whereas robustness corresponds to local damage states. Based on the two distinct levels, it is argued herein
that the existence of redundancy assists in: (1) enhancing the safety margin/reliability of a structure in its intact state;
and (2) mitigating the sensitivity/vulnerability of the structure to localised damage under an accidental situation.
Thirdly, it classifies major aspects of ensuring redundant structures, taking into account various stages they are
involved in. Further, essential means of designing for redundancy are discussed which can readily be integrated into
safety decision-making. A hypothetical structural system is employed for illustrative purpose, indicating that the
aforementioned means can be treated within the reliability and robustness frame. Based on this study, some
recommendations are presented and areas of further research indicated.

© 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.

Keywords: Redundancy; Structural systems; Safety, robustness; Alternate load paths

1. Introduction

Concerning redundancy, a general consensus among the structural engineering community is that it
serves as an important, usually beneficial, factor to structural safety. Extensive effort has been made in

* Corresponding author
E-mail address: fangzhao_xin@yahoo.com.cn

1877–7058 © 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
doi:10.1016/j.proeng.2011.07.273
Z.X. FANG and H.T. FAN / Procedia Engineering 14 (2011) 2172–2178 2173

redundancy analysis, especially in its quantitative appraisal (Frangopol and Curley 1987; Frangopol et al.
1992). Despite this, however, redundancy by itself remains an obscure terminology and still leaves some
room for interpretation and clarification.
Up until the last 10-odd years, the tragic collapse of the Alfred P. Murrah Federal Building and the
twin towers of the World Trade Center invoked renewed interest in safety decision-making for structural
systems when confronted with accidental scenarios. It, therefore, becomes imperative to clarify and
deepen redundancy’s role in enhancing structural robustness/resilience under accidental local damage
(Fang and Li 2007; Fang and Li 2009).
We attempt to set up here a framework for redundancy explanation and realisation. After clarifying
related terms, we provided a working definition for redundancy. With a clear understanding of
redundancy and its ‘leverage effect’ on structural safety, the framework ought to potentially promote
safety decision-making. Further, the exact role of redundancy in ensuring structural safety was explored.
Section 4 focused on how to ensure structural redundancy. Section 5 provided some recommendations
and final remarks.

2. Definitions of indeterminacy and alternative load paths

2.1 Static indeterminacy and alternative load paths

Static indeterminacy rests with the form/configuration and connectivity of a structure in the pristine
state, irrespective of its response to specific loadings. It merely refers to the geometric stability of
structures. As known in classical structural analysis, the degree of static indeterminate equals the
difference between the numbers of unknown reaction forces and independent equilibrium equations.
Alternative load paths mean multiple load paths that enable a redistribution of loads/actions
previously carried by damaged/failed portions. Given that alternate load paths are available, a structure
will have a chance to arrest damage propagation, and widespread collapse may be delayed as a
consequence.

2.2 Redundancy

In civil engineering, redundancy is a frontier concept. The terms redundancy and alternative load path
are often regarded as synonyms, particularly in situations wherein the issues of accidental events and
progressive collapse become apparent (Starossek 2007; Marhadi and Venkataraman 2009). Redundancy
typically stresses the ability of a structural system to redistribute among its members/connections the
loads which can no longer be carried by some other damaged portions (Biondini et al. 2008). To
guarantee redundant structures, the availability of alternative load paths (or additional load-transfer
mechanisms) is of paramount importance. Unlike structures exhibiting redundancy, non-redundant
structures may fail immediately under local damage, such as loss of load carrying element(s). And for this
very reason, redundancy means and also provides an availability of warning prior to system failure
(Okasha and Frangopol 2010).
As can be extracted from the foregoing discussion that, apart from the pristine state, redundancy
stresses the desired global behaviour of a structure under accidental scenarios. Taking this fact into
consideration, we conclude herein that: redundancy is an essential property of a structural system
concerning existence and subsequent development of alternative load path(s) or multiple load-transfer
mechanisms. Owing to the wide variability of accidental scenarios, redundancy shall nominally be viewed
as a property of a structural system alone and being independent of the possible accidental scenarios.
2174 Z.X. FANG and H.T. FAN / Procedia Engineering 14 (2011) 2172–2178

Since static indeterminacy pertains solely to the form of a structure in its pristine state, it is not
interchangeable with redundancy. This argument indirectly explains why an increase in the degree of
static indeterminacy does not necessarily yield an increase of safety margin.

3. Redundancy and structural safety

As regards the influence of redundancy, pilot studies customarily paid attention to its positive
significance, and appraisal indexes as well, in terms of residual strength, failure probability, or reliability
(Frangopol and Curley 1987; Frangopol et al. 1992; Berteto and Berteto 1999). Nevertheless, practical
experience as well as several benchmark studies has shown that: (1) redundancy’s role in an intact
structure can be rather limited (Sebastian 2004; Schafer and Bajpai 2005); (2) increased degree of
redundancy may bring with it increased uncertainty (Elms 2004). By contrast, little effort has been
dedicated to obtain holistic understanding of the redundancy’s role, among which Gorman (1984) pointed
out that redundancy may reduce structural sensitivity to abnormal loadings, but improve structural
reliability under design loads. This statement provides insights into redundancy, though it does not take
note of the distinct working situations involved in structural safety.
From the life-span point of view, structural safety consists of two levels: firstly, a structure shall
withstand loads reliably under normal situations and pristine state; secondly, the structure, or a major part
of it, shall remain stable and avoid disproportionate failure under accidental situations and local damage
state (SCOSS 1994; Fang and Li 2009).

Figure 1: Different levels of structural safety, adapted from Fang (2009).

See Figure 1 for reference, the first level corresponds to the pristine or intact state of a structure
together with its reliability requirement, as covered in current reliability-based design/analysis. The
second level, however, deals with local damage state and robustness requirement which used to be
ignored within the conventional design/analysis envelope. It is thereby evident that principle differences
exist between these levels, and this background implies that the role of redundancy ought to be explored
in concert with the two underlying safety levels.
Within the present scenario, the significance of redundancy remains twofold. At the reliability level,
redundancy (i.e. the existence of alternative load paths) contributes to the safety margin of a structure in
its intact state; at the robustness level, however, redundancy (i.e. the development of alternative load
paths) assists in mitigating the sensitivity/vulnerability of the structure to accidental scenarios. On the
other hand, note that structural robustness, otherwise known as ‘structural integrity’ in North America,
highlights the tolerance of a structural system to accidental scenarios. The requirement of robustness thus
coincides with the objective of hazard mitigation in the sense that “under accidental scenarios, the risk of
Z.X. FANG and H.T. FAN / Procedia Engineering 14 (2011) 2172–2178 2175

disproportionate failure should be mitigated to an acceptably low extent”. Undoubtedly, redundancy


serves as a key factor for structural robustness because it offers the possibility of avoiding an
unacceptable failure (e.g., cascading or progressive collapse) by means of alternative load paths.

4. Means to improve the redundancy of structural systems

4.1 Design of redundant structures

Structural redundancy relies on a variety of factors, including structural form/configuration, member


sizes, material properties, member/connection properties (e.g., resistance, deformation, or ductility),
applied loads and load sequence, and so on (Frangopol and Curley 1987). Designing for redundancy, in
principle, should take these factors into account. The point lies in deciding when and how to incorporate
them into practice. Major aspects of ensuring redundancy will be discussed herein based on different
phases of the life cycle of a structure.

Table 1: Consideration of redundancy enhancement in different phases

Consider four phases, i.e. conceptual design, detailing design, construction, and use/maintenance.
Table 1 has sought to indicate the identified major aspects of ensuring redundancy together with
associated phases. Note that a single circle denotes medium level of correlation whilst two circles denote
high correlation. By way of analogy, structural continuity needs to be addressed in conceptual design, for
example, by setting settlement and aseismic joints rationally; in course of construction, however, the key
to continuity lies in analysing where to set concrete construction joints.
It can be inferred from Section 2.2 that redundancy-based safety decision-making shall/must centre on
alternative loads paths. For general structural systems, redundancy realisation may be conducted by
consideration of the following means: (1) ensuring hyperstatic vertical and/or lateral load-bearing systems;
(2) employing ductile materials and connections; (3) providing resistance to load/moment reversals, for
example, by setting continuous/symmetrical reinforcement through beam to column connections; (4)
providing sufficient tying of structural elements (e.g., edge beams and adjoining slabs).
Recently, Ye (2009) suggested distinguishing between local failure and global failure mechanisms of
different structural systems. This way inspires novel insights into structural system effect which has not
been fully brought to light in conventional design/analysis.
Finally, it is important to point out that redundancy is not always necessary for improvement of the
safety of structural systems. There are certain situations where redundancy does not guarantee an
appropriate robustness, and damage propagation needs to be prevented effectively by means of weak links,
a strategy analogous to the ‘segmentation’ as proposed in (Starossek 2007). Under the circumstances,
reduced continuity and non-ductile failure modes may be more desirable.
2176 Z.X. FANG and H.T. FAN / Procedia Engineering 14 (2011) 2172–2178

4.2 Example

This section provides case study of a structural system and discusses how redundancy is implemented
in the reliability and robustness context.

(a) (b)

(c)

Figure 2: Transition of load-transfer mechanisms: (a) the prototype structure; (b) flexural load transfer mechanism; (c) tensile load
transfer mechanism (catenary action).

Consider an isolated body as shown in Figure 2a, assuming that it was extracted from a typical
concrete frame with strong column/weak beam design. Upon removal of the centre column at the lower
floor, the part which it supported in the pristine frame begins to deflect downward under vertical load as
the form of the frame changes. The damaged structure will absorb the work done by external loads
through the flexural deformation of beams until plastic hinges develop at the beam ends, see Figure 2b. It
is noteworthy that beam AB and BC undergo moment reversals in the vicinity of node B. As damage
propagates and plastic hinges turn to ordinary hinges, the damaging structure may continue to carry
external loads by virtue of catenary action, see Figure 2c. The whole process represents an ideal and
somewhat desirable failure mode for common moment resisting frames: in flexural load transfer
mechanism, the redundancy of the frame depends to a large extent on the ductility (or energy-absorbing
capability) of the beams/nodes; in tensile load transfer mechanism (herein referring to the catenary action),
redundancy requires effective anchorage, e.g., in beam to column connections (i.e. continuity).
As far as the present case is concerned, the ductility of certain beam elements can be achieved by
improving their load-bearing and deformation capacity, including residual capacity. Similar treatments
are usually part of conventional reliability-based design. After local damage takes place, redundancy can
be fulfilled by providing moment reversals resistance and catenary effect. In this situation, the structure is
capable of redistributing the loads among its working components further; in other words, these
performances reduce structural sensitivity to local damage, so that the robustness of the structure is
improved.
Z.X. FANG and H.T. FAN / Procedia Engineering 14 (2011) 2172–2178 2177

5. Conclusion and final remarks

(1) Redundancy was defined here as an essential property of a structural system with regard to existence
and subsequent development of alternate load paths. Hence redundancy refers to the pristine state of
a structure, and, at the same time, stresses its desired global behaviour under accidental scenarios.
(2) The authors investigated the influence of redundancy based on the two safety levels, which
correspond to distinct working/design conditions. It is shown that redundancy helps to enhance the
safety margin of a structure in its intact state and mitigate the sensitivity of the structure to local
damage, i.e. ensure structural robustness. We believe that this explanation creates a bridge between
redundancy, and safety decision-making and hazard mitigation.
(3) Major factors and aspects of redundancy ‘design’ were discussed herein according to various stages
of structural life cycle.
Additionally, there are signs that keen attention will be focused on safety decision-making in the face
of increasingly outstanding natural- and human-induced hazards. The article also exposed some questions
where more research and development is yet required.
(1) What are the primary performance requirements and compliance criteria for the redundancy of
various types of structural systems (e.g., masonry structure, frame-shear wall structure)?
(2) What are the relationships, qualitative or quantitative, between the levels of redundancy and
reliability/robustness? Then how redundant is redundant enough?

Acknowledgments

The support from (1) the Project of Shandong Province Higher Educational Science and Technology
Program under Grant No. J09LE02, and (2) the Startup Foundation for Doctors from Yantai University
under Project No. TM09B1 is gratefully acknowledged.

References

[1] Berteto RD and Berteto VV. Redundancy in earthquake-resistant design. ASCE Journal of Structural Engineering;1999,
125(1), pp. 81-88.
[2] Biondini F, Frangopol DM, and Restelli S. On structural robustness, redundancy and static indeterminacy. Proceedings of the
2008 ASCE-SEI Structures Congress ;2008, Vancouver, pp. 1-10.
[3] Elms DG. Structural safety—issues and progress. Progress in Structural Engineering and Materials; 2004, 6(2), pp. 116-126.
[4] Fang ZX. Problems and prospects of structural robustness analysis. Proceedings of the International Symposium on
Innovation & Sustainability of Structures in Civil Engineering 2009 (Vol. 2), Guangzhou pp. 1703-1707.
[5] Fang ZX and Li HQ. Structural robustness and risk mitigation. Engineering Mechanics, 24(Suppl. 1) ;2007, pp.79-82. (in
Chinese)
[6] Fang ZX and Li HQ Robustness of engineering structures and its role in risk mitigation. Civil Engineering and
Environmental Systems ;2009, 26(3), pp. 223-230.
[7] Frangopol DM and Curley JP. Effects of damage and redundancy on structural reliability. ASCE Journal of Structural
Engineering ;1987,113(7), pp. 1533-1549.
[8] Frangopol DM, Iizuka M, and Yoschida K. Redundancy measures for design and evaluation of structural systems. Journal of
Offshore Mechanics and Arctic Engineering, ASME Transactions; 1992, 114(4), pp. 285-290.
[9] Gorman MR. Structural redundancy. Proceedings of the 4th ASCE Specialty Conference on Probabilistic Mechanics and
Structural Reliability, Berkeley; 1984, pp. 45-48.
[10] Marhadi K and Venkataraman S. Surrogate measures to optimize structures for robust and predictable progressive failure.
Structural and Multidisciplinary Optimization;2009, 39, pp. 245-261.
2178 Z.X. FANG and H.T. FAN / Procedia Engineering 14 (2011) 2172–2178

[11] Okasha NM and Frangopol DM. Redundancy of structural systems with and without maintenance: An approach based on
lifetime functions. Reliability Engineering and System Safety ;2010, 95(5), pp. 520-533.
[12] Schafer BW, Bajpai P. Stability degradation and redundancy in damaged structures. Engineering Structures; 2005, 27(11), pp.
1642-1651.
[13] Sebastian WM. Collapse considerations and electrical analogies for statically indeterminate structures. ASCE Journal of
Structural Engineering; 2004, 130(10), pp. 1445-1453.
[14] Starossek U. Disproportionate collapse: a pragmatic approach. Structures & Buildings; 2007, 160(SB6), pp. 317-325.
[15] The Standing Committee on Structural Safety (SCOSS). 10th Report of SCOSS. SETO Ltd, London; 1994
[16] Ye LP. Failure mechanism and its control of building structures under earthquakes based on system concept. Earthquake
Resistant Engineering and Retrofitting;2009,31(5), pp. 1-7.(in Chinese)

You might also like